Most modern accelerator-based applications require a high-brightness, high-peak and high-average current electron beam that has a short bunch, a small energy spread, and a long lifetime. This type of beam is often referred to as a “super beam.” Photocathodes are typically used to generate such a super beam, because photocathodes have a very high current density capability, compared to thermionic cathodes, and because photocathodes are able to generate short bunched beams that can be matched into RF (radiofrequency) accelerators.
Photocathodes have a limited lifetime, even if operated at ultra-high vacuum (UHV) environments. The main reason is ion-bombardment that occurs during operation. State-of-the-art non-polarized photocathodes can operate at average currents of tens of mA, with a charge lifetime of less than 1000 Coulombs, which is equivalent to a lifetime of less than 28 hours at an average current of 10 mA. State-of-the-art polarized beams can operate at average currents of a few mA with charge lifetime of about 200 Coulombs, which is equivalent to a charge lifetime of less than 6 hours at an average current of 10 mA.
Most modern accelerator projects require electron sources with much higher average currents (with a reasonable operating period), compared to what is provided by the state of the art. In addition, these projects also require low emittance, high-peak currents with short bunch lengths, and small energy spread beams, as required by a super beam.
Apart from charge lifetime issues, many technical challenges remain in developing such a super beam. The requirements of a super beam include a small emission area, a high bunch charge, and a high repetition rate. Also, the beam must be compressed for a short bunch beam. Reducing the emission area and increasing the bunch charge will increase the space charge on the cathode, so that eventually the source becomes space-charge limited.
Attempts to overcome space charge effects on photocathodes include various methods of increasing the electric field gradient on the cathode (ECth), combined with a certain degree of bunch lengthening on the cathode. Methods for pushing up ECth include without limitation: increasing the anode-cathode voltage or reducing the accelerating gap, when DC voltage guns are used; and using RF cavity guns (also referred to as RF guns) or even superconducting RF guns.
As ECth is finite, one has to increase the bunch length on the photocathode, to further reduce the space charge effects on cathode, and compress the bunch later on. Increasing bunch length on the photocathode makes the later beam compressing more difficult, however.
Usually the beam bunch is too long after the beam comes out from gun. This requires the beam bunch to be compressed in order to reduce bunch length. Generally, compression techniques include without limitation: ballistic compression techniques, which utilize the velocity difference in a bunch; and “dispersion optics” compression techniques, which utilize the path length difference of a bunch through a beam line.
In ballistic compression, a beam bunch energy distribution is modified by an RF cavity (bunching cavity) such that the head beam energy is lower than the tail beam energy, then the tail beam catches up with the head beam during drifting and eventually the beam is compressed. This technique requires that the beam not be too relativistic, or else the drifting length would be very long. Assume a beam has a center energy of 1 MeV and head-tail energy difference of ±5% (head energy<tail energy), the bunch length can be compressed 8.0 mm per 1 m of drift.
In dispersion optics compression, the beam bunch energy distribution still needs to be modified in the same way as with ballistic compression. The beam then enters a dispersive beam line such as a symmetric magnetic chicane.
FIG. 1 illustrates a symmetric magnetic chicane that can be used in conventional dispersion optics compression. As the head beam with lower energy travels for a longer distance, compared to the distance traveled by a tail beam with higher energy, the bunch is compressed after the beam passes through the chicane.
Chicane compression is generally suitable for relativistic beams. The compression length (Δs) by a magnetic chicane, at small energy spread (δ), is approximately:
            Δ      ⁢                          ⁢      s        =          2      ⁢      a      ⁢                          ⁢      θ      ⁢                          ⁢                        sin          ⁢                                          ⁢          θ                                      cos            2                    ⁢          θ                    ⁢      δ        ,
where a is the projected distance between the 1st magnet (101) and the 2nd magnet 102 (or between the 3rd magnet 103 and the 4th magnet 104), and θ is the bending angle of each dipole. Assuming that a=1 m and θ=30°, the center energy is 5 MeV and the head-tail energy difference is ±1%, which is the same energy magnitude difference compared to the ballistic compression example. The compression length is about 14 mm through the system.
Both ballistic compression and chicane compression will degrade beam emittance and leave a high final energy spread.
Despite past efforts to overcome space-charge effects, including the compression techniques described above, many challenges thus remain.